{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T08:47:08Z","timestamp":1771922828610,"version":"3.50.1"},"reference-count":46,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T00:00:00Z","timestamp":1769817600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T00:00:00Z","timestamp":1769817600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,31]]},"DOI":"10.1109\/cgo68049.2026.11394849","type":"proceedings-article","created":{"date-parts":[[2026,2,23]],"date-time":"2026-02-23T20:46:32Z","timestamp":1771879592000},"page":"255-267","source":"Crossref","is-referenced-by-count":0,"title":["Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References"],"prefix":"10.1109","author":[{"given":"Hongzheng","family":"Chen","sequence":"first","affiliation":[{"name":"Cornell University,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bin","family":"Fan","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alexander","family":"Collins","sequence":"additional","affiliation":[{"name":"NVIDIA,United Kingdom"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bastian","family":"Hagedorn","sequence":"additional","affiliation":[{"name":"NVIDIA,Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Evghenii","family":"Gaburov","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Masahiro","family":"Masuda","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Matthew","family":"Brookhart","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chris","family":"Sullivan","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jason","family":"Knight","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhiru","family":"Zhang","sequence":"additional","affiliation":[{"name":"Cornell University,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vinod","family":"Grover","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2024.3373763"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3656177"},{"key":"ref3","first-page":"606","article-title":"Efficiently scaling transformer inference","volume-title":"Proceedings of machine learning and systems","volume":"5","author":"Pope"},{"key":"ref4","first-page":"711","article-title":"Data movement is all you need: A case study on optimizing transformers","volume-title":"Proceedings of Machine Learning and Systems","volume":"3","author":"Ivanov"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640399"},{"key":"ref6","article-title":"Nvidia hopper architecture","year":"2022"},{"key":"ref7","article-title":"Nvidia blackwell architecture technical brief","year":"2025"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3315508.3329973"},{"key":"ref9","article-title":"Cutlass","year":"2025"},{"key":"ref10","article-title":"Thunderkittens: Simple, fast, and adorable kernels","volume-title":"The Thirteenth International Conference on Learning Representations","author":"Spector"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2193"},{"key":"ref12","article-title":"Flashinfer: Efficient and customizable attention engine for LLM inference serving","author":"Ye","year":"2025"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/1365490.1365500"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.5555\/3291168.3291211"},{"key":"ref15","article-title":"Ansor: Generating high-performance tensor programs for deep learning","volume-title":"Proceedings of the 14th USENIX Conference on Operating Systems Design and Implementation","author":"Zheng"},{"key":"ref16","article-title":"Tensor program optimization with probabilistic programs","author":"Shao","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3676641.3716249"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3410463.3414632"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3582016.3582018"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3582514.3582517"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/2063384.2063400"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/2555243.2555258"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA57654.2024.00086"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/cgo51591.2021.9370308"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3729262"},{"key":"ref26","article-title":"Gluon","year":"2025"},{"key":"ref27","article-title":"Tilelang: A composable tiled programming model for ai systems","author":"Wang","year":"2025"},{"key":"ref28","article-title":"Pipethreader: software-defined pipelining for efficient dnn execution","volume-title":"Proceedings of the 19th USENIX Conference on Operating Systems Design and Implementation","author":"Cheng"},{"key":"ref29","first-page":"172","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume-title":"Proceedings of the 33rd International Conference on Neural Information Processing Systems","author":"Paszke"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640366"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3771775.3786277"},{"key":"ref32","article-title":"OpenCL 2.0 reference guide","author":"Groups","year":"2015"},{"key":"ref33","article-title":"oneAPI toolkits: Pipe","year":"2023"},{"key":"ref34","article-title":"Vitis hls stream library","year":"2025"},{"key":"ref35","article-title":"Dato: A task-based programming model for dataflow accelerators","author":"Fang","year":"2025"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3656401"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/3490422.3502369"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref39","article-title":"cuBLAS: Basic linear algebra on NVIDIA GPUs","year":"2025"},{"key":"ref40","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"ref41","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref42","article-title":"Grok 2","year":"2025"},{"key":"ref43","article-title":"Qwen3 technical report","author":"Yang","year":"2025"},{"key":"ref44","article-title":"Deepseek-v3 technical report","year":"2025"},{"key":"ref45","article-title":"The Llama 4 herd: The beginning of a new era of natively multimodal ai innovation","year":"2025"},{"key":"ref46","article-title":"Flashattention-2: Faster attention with better parallelism and work partitioning","author":"Dao","year":"2023"}],"event":{"name":"2026 IEEE\/ACM International Symposium on Code Generation and Optimization (CGO)","location":"Sydney, Australia","start":{"date-parts":[[2026,1,31]]},"end":{"date-parts":[[2026,2,4]]}},"container-title":["2026 IEEE\/ACM International Symposium on Code Generation and Optimization (CGO)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11395173\/11394837\/11394849.pdf?arnumber=11394849","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T07:42:00Z","timestamp":1771918920000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11394849\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,31]]},"references-count":46,"URL":"https:\/\/doi.org\/10.1109\/cgo68049.2026.11394849","relation":{},"subject":[],"published":{"date-parts":[[2026,1,31]]}}}